Joint Blog Series on International Law and Armed Conflict: Are Sieges Prohibited under Contemporary IHL?

Editor’s Note: This post the final post in the joint series hosted by the ICRC Humanitarian Law & Policy Blog, EJIL Talk! and Lawfare, and arising out of the 6th Transatlantic Workshop on International Law and Armed Conflict held at the European University Institute in Florence in July.

Contemporary armed conflicts in Syria, Yemen or Iraq have seen a resurgence of sieges of cities and other densely populated areas. This ancient—some would say archaic or medieval—method of warfare is expected to be increasingly used in future urban conflicts. The catastrophic humanitarian consequences of recent prolonged sieges—such as in Ghouta (Syria), where civilians are starving because of lack of access to objects indispensable to their survival—have led to widespread condemnations by the international community (see e.g., UN Security Council Resolution 2139 (2014)). The question discussed here is what does international humanitarian law (IHL) say about siege warfare? Is it explicitly ruled out? How might IHL rules and principles constrain siege warfare?

No explicit IHL rules against siege warfare

Sieges are not per se an explicitly prohibited method of warfare under IHL. On the contrary, one could say that IHL implicitly allow sieges by merely mentioning steps to be taken to mitigate their negative effects on civilians and civilian objects (Art 27 1907 Hague Regulations; Art 15 GCI; Art 18 GCII; Art 17 GCIV).

Sieges have been used throughout history and military doctrine usually regards sieges as essential to the effective conduct of hostilities in order to control a defended locality and obtain surrender or otherwise defeat the enemy through isolation. Since sieges are a harsh method of warfare and are based on almost complete isolation of the besieged locality, their use will almost inevitably involve frictions with numerous rules and principles of IHL—at least when the besieged area involves civilian presence.

Numerous constraints on siege warfare

There are a number of IHL prohibitions that may constrain siege warfare. These include the prohibition against terrorizing the civilian population (Art 51(2) API; Art 13(2) APII; CIHL Study, Rule 2), the prohibition of collective punishment (Art 75 API; Art 4 APII; CIHL Study, Rule 103) or the prohibition of human shields (Art. 51(5) API; CIHL Study, Rule 97). The most obvious prohibition that impacts siege warfare, however, is the prohibition of starvation of civilians (Art 54(1) API; Art 14 APII).There is also a question as to whether conduct of hostilities rules, and the principle of proportionality in particular, may serve as an additional constrain on siege warfare Article 51(5)(b) of API CIHL Study, Rule 14). The present blog post will focus on the latter two—the prohibition against starving civilians, purposefully or incidentally, and the principle of proportionality.  

It is to be noted that IHL also contains norms pertaining to evacuation and humanitarian relief that are of primary importance to protect the civilian population from the effects of siege warfare. But this is the subject of a separate post in this blog series (see Laurie Blank’s post here)

Allowing sieges, prohibiting starvation: Is this really compatible?

The prohibition against the starvation of civilians

Modern IHL prohibits the starvation of civilians—i.e., deliberately depriving them of food—as a method of warfare. Derived from the IHL principle of distinction, this rule appears for the first time in both Additional Protocols of 1977 (Art 54(1) API; Art 14 APII) and, today, is considered customary law in both international and non-international armed conflicts (CIHL Study Rule 53). The Rome Statute provides that ‘intentionally using starvation of civilians as a method of warfare’ is a war crime in international armed conflicts (ICC Statute, Art 8(2)(b)(xxv)), but surprisingly not in non-international armed conflicts. Under domestic criminal law, individuals have also been convicted for the crime of starvation in the context of non-international armed conflicts (see, e.g., Croatia, District Court of Zadar, Perišić and Others case, Judgment, 24 April 1997).

Interpretation of the rule against starvation—the notion of ‘purpose’

A key issue here is whether siege warfare when civilians are present is indirectly prohibited by the prohibition of starvation of civilians since, in practice, civilians will be the first to suffer from deprivations arising out of the siege-induced isolation. What matters here is how the prohibition against starvation is interpreted. In this respect, views vary. The majority view seems to be that sieges are not prohibited even if they cause starvation, as long as their purpose is to achieve a military objective and not to starve the civilian population (see e.g.,  commentary on CIHL Study Rule 53). Some go as far as to require that starvation of civilians is the ‘sole’ or ‘primary’ purpose of a siege to consider it unlawful, which would reduce considerably the value of the provisions on the prohibition of starvation (see San Remo Manual on International Law Applicable to Armed Conflicts at Sea, 12 June 1994, Art 102 a:‘sole purpose’ in relation to sea blockades); Manual on International Law Relating to Air and Missile Warfare, 15 May 2009, Art 157a: ‘sole or primary purpose’ in relation to an aerial blockade’). In practice, it is very difficult, if not impossible, to prove that the ‘purpose’—and particularly the ‘sole or primary purpose’—of a siege is starvation of civilians. However, arguably, failing to attempt an evacuation of civilians, or at least the most vulnerable among them, coupled with a denial of humanitarian assistance should suffice to indicate that the purpose of the siege is to starve civilians.

A combined reading of Articles 54(2) and (3): A prohibition on incidental starvation?

But this should not be the end of the analysis. It is often forgotten that, even if a restrictive interpretation is given to starvation of civilians as a ‘method of warfare’ based on the notion of purpose (Art 54(1) API), the prohibition of acts that have the effect to starve the civilian population, which is to be found in the Additional Protocols, is wider. As a corollary to the prohibition of starvation of civilians, IHL also prohibits attacking, destroying, removing or rendering useless objects indispensable to the survival of the civilian population (e.g., foodstuffs, agricultural areas, crops, livestock, drinking water and irrigation systems) for the specific purpose of denying these items for their sustenance value to the civilian population or to the adverse party, whether in order to starve out civilians, to cause them to move away, or for any other motive. (Art 54(2) API; CIHL Study, Rule 54).

A combined reading of Articles 54(2) and (3) API shows that more than mere purposeful starvation of civilians is prohibited. Take, for instance, the destruction of a drinking water installation in the context of a siege that has the specific purpose of denying water to the adverse Party. This will be considered as unlawful if such destruction ‘may be expected to leave the civilian population with such inadequate food or water as to cause its starvation or force its movement’ (Art 54(3)(b) API). Although Article 14 APII is drafted differently, the ICRC Commentaries provide an interpretation that is similar to the combined reading of Articles 54(2) and (3) of API (see 1987 ICRC Commentary to Art 14 APII, paras 4806-4807).

Incidental starvation is therefore also prohibited under certain (not all) circumstances. Article 54(2) covers only situations where the belligerent party deprives the enemy from existing resources (see ‘attack, destroy, remove or render useless’), not when it prevents the enemy from being resupplied. This subtle reading of Article 54 led eminent scholars to consider that the Additional Protocols considerably restrict the possibility to resort to sieges. In practice, indeed, sieges will almost inevitably have the side effect of starving the civilian population or forcing its movement. Already in 1991, Yoram Dinstein wrote that, with the 1977 prohibition of starvation, ‘a true siege would no longer be feasible’ (see Yoram Dinstein, Siege Warfare and the Starvation of Civilians, in Astrid J.M. Delissen and Gerard J. Tanja, Humanitarian Law of Armed Conflict: Challenges Ahead, pp150-151). This conclusion leads Dinstein to consider that the Additional Protocols are unrealistic because siege warfare is such a valuable method of warfare from a military perspective. Sean Watts is of the view that, under API, a siege is ‘almost entirely prohibited with respect to life-sustaining objects for civilians’. Beth Van Schaack highlights that, in current armed conflicts ‘it [is] very difficult for a commander to conduct a siege that is both successful and lawfulʼ.

Whether this prohibition of sieges causing incidental starvation in the circumstances described above is considered as realistic or not, this is now the current state of the law for States which are parties to the Additional Protocols. What remains unclear is whether customary law goes this far as well.

Does the principle of proportionality constrain siege warfare?

Another question, which proved to be even more controversial at the Transatlantic Workshop on International Law and Armed Conflict is whether the IHL principle of proportionality restricts siege warfare.

Of course, when sieges are conducted through bombardments, the principle of proportionality applies. But consider the following, more complex, situation. A belligerent party is besieging a small defended town by merely preventing weapons, as well as food and other objects indispensable to the survival of the population to enter in the area. This party does comply with Article 54 API. Its purpose is to weaken the enemy and obtain its surrender, not to starve the civilian population. And, the party does not attack, destroy, remove or render useless objects indispensable to the survival of the civilian population. The belligerent party merely isolates the town and prevents it from receiving supplies. After a few weeks, the civilian population has exhausted its food reserves and the most vulnerable start to die from starvation while other civilians attempt to leave the area at the risk of their own lives. After some months, the humanitarian situation becomes unbearable, the human toll among civilians is high while the expected military advantage is much lower than what the besieging party hoped for because the armed enemy controls and continues to use the means of survival that remain available. Shouldn’t we consider the furtherance of the siege as disproportionate from a humanitarian law perspective? Instinctively, the answer should be yes.

Technically speaking, the answer is less straightforward under the law. Article 51(5)(b) of API—which encompasses the IHL principle of proportionality and which reflects to a large extent customary law (CIHL Study, Rule 14)—prohibits attacks which may be expected to cause ‘collateral damage’ which would be excessive in relation to the concrete and direct military advantage anticipated.

The question here is whether a siege can be considered as an ‘attack’. In my view, the answer is yes. I submit three arguments in support of this proposition.

• Article 49 of API defines ‘attacks’ in a sufficiently broad and flexible manner to include siege. It defines ‘attacks’ as ‘acts of violence against the adversary, whether in offence or in defence’. How could starving civilians not be considered as an act of violence?
• To the extent that siege warfare is a method of warfare—or a combination of several methods of warfare (bombing and starvation)—conduct of hostilities rules and principles do apply. As highlighted by James Kraska, ‘siege implicates the principle of distinction and proportionality in the law of armed conflict’. Similarly, Sean Watts is of the view that ‘the required precautions against disproportionate incidental harm are therefore exceedingly important during siege operations’.
• In the context of blockades, with which sieges entertain a number of similarities, nobody has questioned the relevance of the principle of proportionality. Both the San Remo Manual (para 102b) and the Manual on International Law Relating to Air and Missile Warfare (para 157b) provide for it.

Assuming the relevance of the principle of principle of proportionality, the legality of sieges needs to be continually monitored and assessed in light of the concrete and direct military advantage anticipated and expected collateral damages (including number of civilians expected to starve, suffer from malnutrition and other nutritional deficiencies, forced to move etc.) In this context, the duration of the siege (a few weeks or rather years) should be an important consideration.

Perhaps one way to render sieges more humane would be to further investigate how the principle of proportionality operates in the context of siege warfare.